Road congestion evaluation method based on 5S

Abstract

The invention relates to the technical field of intelligent traffic management, in particular to a 5S-based road congestion evaluation method, which comprises the following steps of: determining multi-dimensional indexes from five aspects of fluid property, structural property, spatiality, scenario property and social property elements, collecting the corresponding multi-dimensional indexes, and preprocessing; a fuzzy comprehensive evaluation model is constructed by using the combination of an analytic hierarchy process and fuzzy mathematics, the evaluation precision is improved through weight distribution and a membership function, contrastive analysis is performed in combination with examples, and finally, according to real-time traffic data, each index weight is dynamically adjusted, and the model adaptability of the fuzzy comprehensive evaluation model is improved. The dynamic evaluation method suitable for the complex scene supports real-time data input and multi-road parallel calculation, and the problem that a traditional evaluation method mainly depends on a single index is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent transportation management, and particularly relates to a road congestion evaluation method based on 5S. Background Art

[0002] Currently, the problem of urban traffic congestion is becoming increasingly serious. Traditional evaluation methods mainly rely on single indicators (such as traffic flow, average speed, etc.), and have the following limitations:

[0003] One-sidedness: Single indicators cannot comprehensively reflect the complex causes of traffic congestion (such as road structure, special weather, social perception, etc.);

[0004] Poor adaptability: In special scenarios such as traffic accidents or extreme weather, the evaluation results deviate greatly from the actual congestion situation;

[0005] Lack of systematicness: Existing methods do not integrate the multi-element correlation of the traffic system from a global perspective.

[0006] In the prior art, although methods such as evaluation models based on neural networks and grey system theory have been improved, the problem of multi-dimensional comprehensive evaluation has not been solved yet. Summary of the Invention

[0007] The purpose of the present invention is to provide a road congestion evaluation method based on 5S, combining a comprehensive evaluation method of systematic theory and fuzzy mathematics to improve the accuracy and practicality of traffic congestion analysis.

[0008] To achieve the above purpose, the present invention provides a road congestion evaluation method based on 5S, including the following steps:

[0009] Step 1: Determine multi-dimensional indicators based on the 5S theory;

[0010] Step 2: Collect and preprocess multi-dimensional indicator data;

[0011] Step 3: Use the analytic hierarchy process for weight calculation and consistency test;

[0012] Step 4: Construct a fuzzy comprehensive evaluation model;

[0013] Step 5: Conduct a comparative analysis through comprehensive examples;

[0014] Step 6: Optimize and expand the fuzzy comprehensive evaluation model.

[0015] Optionally, the multi-dimensional indicators in step 1 include traffic flow, density, capacity utilization rate, travel time index TTI, and queue time index QTI. The 5S in the 5S theory consists of Streams, Structures, Spaces, Scenarios, and Societies.

[0016] The traffic flow is the real-time traffic volume data on the road, with the unit of hundreds of vehicles per hour, mapping the Streams in the 5S theory.

[0017] The density is the number of vehicles on the road per unit length, with the unit of vehicles per kilometer, mapping the Structures.

[0018] The capacity utilization rate is calculated by the ratio of the designed capacity of the road to the actual traffic flow, with the unit of percentage, mapping the Spaces.

[0019] The travel time index TTI is the ratio of the actual travel time of vehicles in the real-time traffic condition to the free flow travel time, mapping the Scenarios.

[0020] The queue time index QTI is the queue time of vehicles at the signalized intersection, mapping the Societies.

[0021] Optionally, in step 3, a judgment matrix of the 5S theory elements is constructed according to the expert scoring method, the weight vector is calculated by the eigenvector method, and the consistency ratio is calculated by the consistency index and the random consistency index.

[0022] Optionally, the process of determining the modeling mechanism of the fuzzy comprehensive evaluation model in step 4 includes the following steps:

[0023] Step 4.1: Determine the factor set and comment set of the evaluation object.

[0024] Step 4.2: Determine the weight coefficient distribution matrix of the evaluation factors.

[0025] Step 4.3: Determine the single-factor evaluation matrix.

[0026] Step 4.4: Calculate the fuzzy comprehensive evaluation result matrix.

[0027] Step 4.5: Calculate the comprehensive score of the evaluation object and determine the level of the evaluation object.

[0028] Optionally, the execution process of step 6 includes the following steps:

[0029] Step 6.1: Perform dynamic adjustment based on the multivariate time series analysis of real-time traffic data and the feature recognition technology of different traffic scenarios.

[0030] Step 6.2: Targeted adjustment of the threshold for multi-city applications;

[0031] Step 6.3: Visual display is achieved by developing a traffic congestion evaluation system based on the WebGIS architecture.

[0032] The present invention provides a 5S-based road congestion evaluation method, which determines multi-dimensional indicators from five aspects: fluidity, structure, space, scenario, and social elements, collects the corresponding multi-dimensional indicators for preprocessing, then uses the analytic hierarchy process combined with fuzzy mathematics to construct a fuzzy comprehensive evaluation model, improves the evaluation accuracy through weight assignment and membership functions, conducts comparative analysis with examples, and finally dynamically adjusts the weights of each indicator according to real-time traffic data to enhance the model adaptability of the fuzzy comprehensive evaluation model. The present invention is applicable to dynamic evaluation methods for complex scenarios, supports real-time data input and parallel computing of multiple roads, and improves the problem that traditional evaluation methods mainly rely on a single indicator. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0034] Figure 1 It is a schematic flowchart of the steps of a 5S-based road congestion evaluation method of the present invention.

[0035] Figure 2 It is a schematic flowchart of the construction of the fuzzy comprehensive evaluation model in a specific embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The following details the embodiments of the present invention. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below with reference to the drawings are exemplary and are intended to explain the present invention and should not be construed as a limitation of the present invention.

[0037] The present invention provides a 5S-based road congestion evaluation method, including the following steps:

[0038] Step 1: Determine multi-dimensional indicators based on the 5S theory;

[0039] Step 2: Multi-dimensional indicator data collection and preprocessing;

[0040] Step 3: Use the analytic hierarchy process for weight calculation and consistency test;

[0041] Step 4: Construct a fuzzy comprehensive evaluation model;

[0042] Step 5: Conduct a comparative analysis through comprehensive examples;

[0043] Step 6: Optimize and expand the fuzzy comprehensive evaluation model.

[0044] The detailed step - by - step process is as Figure 1 shown below, and further explanations are provided in combination with the implementation steps:

[0045] The multi - dimensional indicators in Step 1 include traffic flow, density, capacity utilization rate, travel time index (TTI), and queue time index (QTI). The 5S in the 5S theory consists of Streams (fluidity), Structures (structure), Spaces (space), Scenarios (scenario), and Societies (sociality);

[0046] Step 2: Data collection and processing

[0047] 1. Data sources

[0048] Traffic flow: Obtain real - time vehicle flow data through roadside sensors of the traffic management department, Amap API, etc. The unit is hundreds of vehicles per hour.

[0049] Traffic density: Use in - vehicle GPS data and intersection cameras to count the number of vehicles on a unit - length road. The unit is vehicles per kilometer.

[0050] Traffic capacity utilization rate: Calculate through the ratio of the road design capacity to the actual traffic flow. The unit is percentage (%).

[0051] Travel Time Index (TTI): Based on the real - time traffic condition data of Amap, calculate the ratio of the actual travel time of vehicles to the free - flow travel time.

[0052] Queue Time Index (QTI): Through intersection cameras and signal control systems, count the queue time of vehicles at signalized intersections.

[0053] Specifically, in the traffic flow data collection link, the collection frequency is accurately set according to the traffic functional attributes, carrying capacity of the road, and statistical characteristics of historical traffic flow fluctuations. For urban arterial roads that undertake the main commuting and logistics transportation functions, due to their rapid traffic flow changes and significant impact on traffic congestion, a high - frequency real - time collection strategy of once every 5 minutes is adopted to capture the instantaneous changes in vehicle flow; for sub - arterial roads, since their traffic flow is relatively stable, data is collected once every 10 minutes to ensure effective acquisition of traffic flow information while optimizing data collection costs and system resource occupancy.

[0054] 2. Data Preprocessing

[0055] Data cleaning: Remove outliers (such as data anomalies caused by extreme weather or equipment failure).

[0056] Data normalization: Normalize indicator data of different dimensions to the [0,1] interval to facilitate subsequent calculations.

[0057] Data storage: Store the processed data in the database for easy model calling.

[0058] The specific implementation process is as follows:

[0059] Data transmission:

[0060] The collected multi-dimensional data such as traffic flow, traffic density, capacity utilization, travel time index (TTI) and queue time index (QTI) are transmitted in real time and efficiently through wireless communication technology based on cellular network architecture (such as 4G / 5G network). The data is quickly delivered to the data processing center by utilizing the wide-area coverage and high-speed data transmission capabilities of wireless communication. In areas where there are blind spots or severe interference with wireless signals, a wired network based on fiber-optic transmission is enabled as a backup transmission path, and the high bandwidth, low loss and strong anti-interference characteristics of fiber-optic communication are used to ensure the continuity and stability of data transmission.

[0061] Outlier judgment and processing:

[0062] In the data cleaning stage, the determination of outliers follows strict statistical and traffic engineering standards. If the traffic flow exceeds 150% of the peak value of historical statistical data for the same period, or is lower than 30% of the lower limit of normal flow established based on long-term traffic data, it is determined to be abnormal; the traffic density exceeds 120% of the road design carrying density, indicating that the traffic flow state deviates seriously from the normal level and is considered abnormal; the capacity utilization rate is greater than 100% or less than 0, which violates the physical meaning and the actual traffic operation law and is considered abnormal data; the travel time index (TTI) is less than 1 or greater than 5, and the queue time index (QTI) is greater than 2 times the average queue time obtained by the historical big data analysis of the intersection, which are all determined to be outliers. For the identified outliers, the mean substitution method of adjacent time period data based on time series analysis is used for processing. By analyzing the correlation and trend of the data in adjacent time periods, the outliers are replaced with reasonable data values ​​to ensure the accuracy and reliability of the data.

[0063] Normalization:

[0064] Data normalization uses the min-max normalization method based on linear transformation, and its mathematical principle is based on the linear mapping of data space. The specific formula is: Where x is the original data, (x min ) and (xmax ) They are the minimum and maximum values of the indicator within the acquisition time period respectively. By this method, all indicator data are linearly mapped to the interval [0, 1], eliminating the numerical deviation caused by the differences in dimension and magnitude between different indicators, making each indicator have the same influence in the subsequent model calculation, and improving the stability and accuracy of the model.

[0065] Step 3: Weight calculation and consistency test

[0066] Steps of the Analytic Hierarchy Process (AHP)

[0067] Construct a judgment matrix: According to the expert scoring method, construct a judgment matrix of 5S elements.

[0068] Calculate the weight vector: Calculate the weight vector through the eigenvector method

[0069] Consistency test: Calculate the consistency index (CI) and the random consistency ratio (CR) to ensure CR < 0.1.

[0070] The following is an explanation of the specific implementation process:

[0071] Judgment matrix construction: When constructing the judgment matrix of 5S theory elements based on the expert scoring method, an expert team composed of at least 5 senior experts with multi-field and multi-professional backgrounds was formed, including experts with profound theoretical research and practical experience in the field of traffic planning, scholars focusing on traffic engineering technology innovation and application, and professionals long engaged in traffic management and decision-making. Experts used the 1-9 scale method for quantitative scoring based on the relative importance of each indicator's impact on road congestion. This scale method is based on the principles of psychology and decision science. 1 means that two factors have the same importance when compared, reflecting the equivalence of the impact of two factors on congestion in a specific traffic scenario; 3 means that the former is slightly more important than the latter, reflecting a slight impact difference; 5 means that the former is significantly more important than the latter, indicating a significant impact difference; 7 means that the former is strongly more important than the latter, meaning there is a large weight difference between the two in terms of affecting road congestion; 9 means that the former is extremely more important than the latter, representing a very high importance difference. 2, 4, 6, and 8 are the intermediate values of the above adjacent judgments, used to more finely depict the relative importance between factors. Combining the scoring results of each expert, the mean calculation method in statistics is used to take the average value as the elements of the judgment matrix to reduce individual subjective bias and improve the objectivity and reliability of the judgment matrix.

[0072] Weight vector calculation: Calculate the weight vector through the eigenvector method, whose theoretical basis comes from the mathematical properties of matrix eigenvalues and eigenvectors. First, use the square root method to calculate the maximum eigenvalue (λ max ) of the judgment matrix. The specific steps are as follows: First, calculate the product of each row element of the judgment matrix (n is the order of the judgment matrix, (a ij ) is a judgment matrix element), the product reflects the comprehensive reflection of the relative importance of the indicator represented by this row and other indicators; then calculate the nth root of \(M_i\) The product result is converted into a comparable relative weight estimate by square root operation; finally Normalization is performed, that is, The sum of the elements of the weight vector is made to be 1, which conforms to the mathematical definition and practical meaning of the weight. Taking the judgment matrix in the present invention as an example, after the above rigorous mathematical calculation process, the weight vectors of traffic flow, traffic density, traffic capacity utilization, TTI, and queuing time index are obtained as (0.49, 0.06, 0.03, 0.13, 0.29), which accurately reflects the relative importance of each indicator in the evaluation of road congestion.

[0073] The single factor evaluation matrix determines:

[0074] When determining the single-factor evaluation matrix, the five key indicators of flow, density, capacity utilization, TTI, and queuing time index are quantitatively analyzed using the membership function based on their evaluation criteria and the probability distribution characteristics of actual data to determine the degree of membership of each indicator at different evaluation levels. For indicators whose values ​​are smaller, the better, such as flow and traffic density, the descending semi-trapezoidal membership function is used. This function is based on the principle of fuzzy mathematics and can accurately describe the transition process of the indicator value from fully meeting a certain evaluation level to gradually deviating from that level;

[0075]

[0076] For indicators whose values ​​are larger and better (if they exist), the ascending semi-trapezoidal membership function is used.

[0077]

[0078] Among them, a and b are thresholds determined according to the evaluation criteria of each indicator. For example, for the flow indicator, the upper limit of very smooth traffic is 20 vehicles / hour (a=20), and the upper limit of smooth traffic is 35 vehicles / hour (b=35). According to this function, the membership degree of each indicator under different evaluation levels is calculated, and then a single factor evaluation matrix is ​​constructed. This matrix accurately reflects the support degree of each indicator for different congestion evaluation levels, and provides basic data for subsequent fuzzy comprehensive evaluation.

[0079] Step 4: Construct fuzzy comprehensive evaluation model

[0080] Fuzzy comprehensive evaluation modeling mechanism

[0081] (1) Determine the factor set and comment set of the evaluation object:

[0082] When constructing an evaluation system, it is necessary to clarify various factors that affect the results of the evaluation object according to the actual situation, and then form a factor set X. Each factor in the factor set usually has a certain degree of fuzziness. Subsequently, according to specific requirements, the evaluation levels are divided into n levels to determine the corresponding comment set of the evaluation object.

[0083] Factor set: X = {x1, x2, x3, x4, x5}

[0084] Judgment set: Y = {y1, y2, y3, y4, y5}

[0085] (2) Determine the weight coefficient distribution matrix of the evaluation factors:

[0086] W = (W1, W2, W3…Wm) (1)

[0087] According to the normalization principle, it is necessary to meet W1 + W2 + W3 + …Wm = 1

[0088] (3) Determine the single-factor evaluation matrix

[0089]

[0090] (4) Calculate the fuzzy comprehensive evaluation result matrix:

[0091] Perform a δ operation on the obtained W and R to obtain the result matrix B

[0092] (5) Calculate the comprehensive score of the evaluation object and determine the level of the evaluation object.

[0093] 4.1 Determine the weight vector of the evaluation index (SPSSAU, Mpai)

[0094] It is determined that traffic flow, traffic density, traffic capacity utilization rate, TTI, and queue time index represent the evaluation factor set x = {x1, x2, x3, x4, x5} in turn; the five levels of traffic congestion, namely very smooth, smooth, slightly congested, moderately congested, and severely congested, represent the critical set Y = {y1, y2, y3, y4, y5} in turn.

[0095] In order to make the weight ratio scientific and appropriate, in the present invention, the principle of the analytic hierarchy process is used to calculate the weights of the five indicators of traffic flow, parking delay time, traffic capacity utilization rate, TTI, and queue time index. The consistency test process is as follows:

[0096] a) First, calculate the consistency test index CI.

[0097] W1 W2 W3 W4 W5 W1 1 7 9 4 3 W2 1 / 7 1 3 1 / 3 1 / 5 W3 1 / 9 1 / 3 1 1 / 5 1 / 7 W4 1 / 4 3 5 1 1 / 3 W5 1 / 3 5 7 3 1

[0098] b) After passing the consistency test, unitarization is performed using equation (3).

[0099]

[0100] c) The following steps are taken to conduct a consistency check on the obtained judgment matrix:

[0101] Calculate the consistency index CI. By substituting the data, the CI value is obtained as 0.055

[0102] d) According to the RI table, the corresponding RI value is found to be 0.11

[0103] CR = CI / RI = 0.05 < 0.1

[0104] Passing the consistency test, the weight vectors for the traffic flow, traffic density, traffic capacity utilization rate, TTI, and queuing time index are as follows:

[0105] W = (0.49, 0.06, 0.03, 0.13, 0.29) (4)

[0106] The detailed process is as Figure 2 shown.

[0107] Step 5: Conduct a comparative analysis with comprehensive examples

[0108] Taking Mingxiu Road, Chaoyang Road, and Minzu Avenue in Nanning as the research objects, real-time traffic data is collected.

[0109] Input data

[0110] Mingxiu Road: Traffic flow = 5200 vehicles / hour, traffic density = 47 vehicles / km, traffic capacity utilization rate = 62%, TTI = 1.68, queuing time index = 1.87.

[0111] Chaoyang Road: Traffic flow = 6100 vehicles / hour, traffic density = 63 vehicles / km, traffic capacity utilization rate = 89%, TTI = 1.96, queuing time index = 2.23.

[0112] Minzu Avenue: Traffic flow = 5700 vehicles / hour, traffic density = 58 vehicles / km, traffic capacity utilization rate = 81%, TTI = 1.89, queuing time index = 2.06.

[0113] Model output

[0114] Mingxiu Road: Congestion index = 0.62 (mild congestion);

[0115] Chaoyang Road: Congestion index = 0.83 (moderate congestion);

[0116] Minzu Avenue: Congestion index = 0.79 (mild congestion).

[0117] Result comparison

[0118] Compared with the traditional method, the evaluation results of the present invention are highly consistent with the actual traffic conditions, and the error rates are all lower than 2%.

[0119] Step 6: Model optimization and expansion

[0120] Dynamic weight adjustment:

[0121] According to the real-time traffic data, dynamically adjust the weights of each index to improve the adaptability of the model.

[0122] The dynamic weight adjustment is based on the multivariate time series analysis of real-time traffic data and the feature recognition technology of different traffic scenarios. During the morning and evening rush hours on weekdays, the traffic flow shows concentrated and regular changes. At this time, the main influencing factors of traffic congestion are traffic volume and queuing situation. Through the mining and analysis of historical morning and evening rush hour traffic data, a weight adjustment model based on machine learning is established. Increase the weight of the traffic volume index by 20%-30% and the weight of the queuing time index by 10%-20%, and adjust the weights of other indexes accordingly to highlight the leading role of traffic volume and queuing in congestion evaluation. When a traffic accident occurs, the traffic flow state of the accident section changes suddenly, and traffic density and travel time index (TTI) become the key factors affecting congestion. An event-driven weight adjustment algorithm is adopted to increase the weights of traffic density and TTI indexes of the accident section by 30%-50% and reduce the traffic volume weight by 20%-30%, so that the model can quickly adapt to the congestion changes caused by traffic emergencies. The adjustment algorithm uses a neural network model based on adaptive weight adjustment. This model takes real-time traffic data as input and automatically adjusts the dynamic weights of each index through the learning and optimization ability of the neural network to ensure that the model can accurately evaluate the road congestion condition under different traffic scenarios.

[0123] Application in multiple cities:

[0124] By adjusting the threshold parameters, the model is extended to other cities to verify its universality.

[0125] When applied in multiple cities, considering the significant differences in geographical environment, population distribution, transportation infrastructure, and travel habits among different cities, it is necessary to make targeted adjustments to the threshold parameters in the model. Using Geographic Information System (GIS) technology and traffic big data analysis methods, comprehensively analyze the road network structure, traffic flow distribution law, and traffic management policies of the target city. For cities with tight road resources and large traffic flows, according to the urban land use plan and traffic carrying capacity, appropriately reduce the congestion threshold of traffic capacity utilization rate to more sensitively reflect the traffic congestion situation; for cities with significant differences in the travel habits of traffic participants, by analyzing the spatio-temporal distribution characteristics of residents' travel, adjust the judgment threshold of the queuing time index. By collecting and analyzing at least three months of historical traffic data of the target city, combined with the expert experience knowledge in the traffic field, using a data-driven parameter optimization algorithm, determine the threshold parameters suitable for this city, and verify the universality and effectiveness of the model in different urban traffic environments.

[0126] Visualization display:

[0127] Develop a traffic congestion evaluation system to display the evaluation results in real time and support the decision-making of traffic management departments.

[0128] The visualization display is realized by developing a traffic congestion evaluation system based on the WebGIS architecture, which integrates the spatial analysis ability of the geographic information system and the interactivity of Web technology. Use an advanced map rendering engine to present the road network in a high-precision and intuitive map form, and mark sections with different congestion levels in different colors, following the international common traffic congestion visualization standards, such as green for very smooth sections, yellow for slightly congested sections, orange for moderately congested sections, and red for severely congested sections, enabling traffic management personnel to quickly identify congested areas. At the same time, set interactive information pop-ups on the map, and use front-end development technology to realize that clicking on a section can display detailed information such as the real-time traffic flow, traffic density, and congestion index of that section, providing users with a rich traffic information query function. The system also integrates data visualization components to provide a historical congestion data query function, and displays the congestion change trend of specific sections at different time periods in the form of professional data visualization charts such as line charts and bar charts. Through the analysis and mining of historical data, it provides an intuitive and effective basis for traffic management departments to make scientific traffic decisions.

[0129] In summary, the present invention has the following beneficial effects:

[0130] Comprehensiveness: Comprehensively quantify traffic congestion from five dimensions of the 5S theory, covering elements of fluidity, structure, space, scenario, and sociality;

[0131] High precision: Combining fuzzy mathematics and the analytic hierarchy process significantly reduces the evaluation error (error rate ≤ 2%);

[0132] Strong adaptability: applicable to different urban roads and special scenarios (such as traffic accidents, bad weather);

[0133] Decision support: providing real-time and dynamic congestion evaluation data for traffic management departments to support the formulation of precise traffic diversion strategies.

[0134] The above-disclosed are only one or more preferred embodiments of the present invention. Certainly, the scope of the rights of the present invention cannot be limited thereby. Those of ordinary skill in the art can understand all or part of the processes of implementing the above embodiments, and the equivalent changes made according to the claims of the present invention still fall within the scope covered by the invention.

Claims

1. A road congestion evaluation method based on 5S, characterized in that It includes the following steps: Step 1: Determine multi-dimensional indicators based on the 5S theory; Step 2: Collect and preprocess multi-dimensional indicator data; Step 3: Use the analytic hierarchy process for weight calculation and consistency test; Step 4: Construct a fuzzy comprehensive evaluation model; Step 5: Conduct a comparative analysis with comprehensive examples; Step 6: Optimize and expand the fuzzy comprehensive evaluation model.

2. The 5S-based road congestion evaluation method according to claim 1, characterized in that the multi-dimensional indicators in Step 1 include traffic flow, density, capacity utilization rate, travel time index TTI, and queue time index QTI. The 5S in the 5S theory consists of Streams, Structures, Spaces, Scenarios, and Societies; the traffic flow is the real-time traffic volume data on the road, with the unit of hundreds of vehicles per hour, mapping to the Streams in the 5S theory; the density is the number of vehicles per unit length of the road, with the unit of vehicles per kilometer, mapping to the Structures; the capacity utilization rate is calculated by the ratio of the designed road capacity to the actual traffic flow, with the unit of percentage, mapping to the Spaces; the travel time index TTI is the ratio of the actual travel time of vehicles in real-time road conditions to the free flow travel time, mapping to the Scenarios; the queue time index QTI is the queue time of vehicles at signalized intersections, mapping to the Societies.

3. The 5S-based road congestion evaluation method according to claim 2, characterized in that in Step 3, a judgment matrix of the 5S theory elements is constructed according to the expert scoring method, the weight vector is calculated by the eigenvector method, and the consistency ratio is calculated by the consistency index and the random consistency index.

4. The 5S-based road congestion evaluation method according to claim 3, characterized in that the process of determining the modeling mechanism of the fuzzy comprehensive evaluation model in Step 4 includes the following steps: Step 4.1: Determine the factor set and comment set of the evaluation object; Step 4.2: Determine the weight coefficient distribution matrix of the evaluation factors; Step 4.3: Determine the single-factor evaluation matrix; Step 4.4: Calculate the fuzzy comprehensive evaluation result matrix; Step 4.5: Calculate the comprehensive score of the evaluation object and determine the level of the evaluation object.

5. The 5S-based road congestion evaluation method according to claim 4, characterized in that the execution process of Step 6 includes the following steps: Step 6.1: Conduct dynamic adjustment based on the multivariate time series analysis of real-time traffic data and the feature recognition technology of different traffic scenarios; Step 6.2: Make targeted adjustments to the application thresholds for multiple cities; Step 6.3: Achieve visual display by developing a traffic congestion evaluation system based on the WebGIS architecture.

Images